analytics-tracking
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to read and incorporate content from local project files (.agents/product-marketing-context.md or .claude/product-marketing-context.md) to guide its responses. This creates a surface for indirect prompt injection, as these files could be modified by a user or external scripts to include malicious instructions that influence the agent's behavior.
- Ingestion points: .agents/product-marketing-context.md, .claude/product-marketing-context.md (as specified in SKILL.md).
- Boundary markers: None present; the skill directs the agent to read the file content directly without specific delimiters or instructions to ignore embedded commands.
- Capability inventory: The skill itself does not define dangerous tools or scripts. However, the agent's inherent capabilities (e.g., file system or network access) could be targeted by instructions contained within the context files.
- Sanitization: No sanitization or validation is performed on the ingested content from the context files.
Audit Metadata